smids_3x_deit_base_sgd_00001_fold4

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.0826
• Accuracy: 0.4333

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1201	1.0	225	1.1112	0.3333
1.1056	2.0	450	1.1099	0.34
1.0987	3.0	675	1.1086	0.3433
1.1099	4.0	900	1.1074	0.355
1.0994	5.0	1125	1.1062	0.3517
1.106	6.0	1350	1.1051	0.3583
1.1031	7.0	1575	1.1040	0.3633
1.1065	8.0	1800	1.1029	0.37
1.0902	9.0	2025	1.1018	0.3683
1.0803	10.0	2250	1.1008	0.3717
1.0894	11.0	2475	1.0998	0.375
1.095	12.0	2700	1.0989	0.3817
1.0882	13.0	2925	1.0979	0.3867
1.0908	14.0	3150	1.0971	0.39
1.1022	15.0	3375	1.0962	0.3917
1.0922	16.0	3600	1.0954	0.395
1.0943	17.0	3825	1.0946	0.3967
1.0851	18.0	4050	1.0938	0.4017
1.0874	19.0	4275	1.0931	0.405
1.0966	20.0	4500	1.0924	0.4083
1.0868	21.0	4725	1.0917	0.4083
1.0765	22.0	4950	1.0910	0.4083
1.0918	23.0	5175	1.0904	0.41
1.0777	24.0	5400	1.0898	0.4183
1.0939	25.0	5625	1.0892	0.42
1.0798	26.0	5850	1.0886	0.4217
1.0858	27.0	6075	1.0881	0.425
1.061	28.0	6300	1.0876	0.4233
1.083	29.0	6525	1.0871	0.425
1.0868	30.0	6750	1.0867	0.425
1.0886	31.0	6975	1.0862	0.4267
1.0841	32.0	7200	1.0858	0.4267
1.0853	33.0	7425	1.0855	0.4283
1.0704	34.0	7650	1.0851	0.4283
1.0702	35.0	7875	1.0848	0.4267
1.0848	36.0	8100	1.0845	0.4283
1.0671	37.0	8325	1.0842	0.4283
1.0578	38.0	8550	1.0840	0.43
1.0817	39.0	8775	1.0837	0.43
1.0866	40.0	9000	1.0835	0.4317
1.083	41.0	9225	1.0833	0.4333
1.0747	42.0	9450	1.0832	0.4333
1.0816	43.0	9675	1.0830	0.4333
1.0657	44.0	9900	1.0829	0.4333
1.0619	45.0	10125	1.0828	0.4333
1.067	46.0	10350	1.0827	0.4333
1.0593	47.0	10575	1.0827	0.4333
1.0587	48.0	10800	1.0826	0.4333
1.0675	49.0	11025	1.0826	0.4333
1.0632	50.0	11250	1.0826	0.4333

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2